EP3438933B1 - Environment recognition device - Google Patents
Environment recognition device Download PDFInfo
- Publication number
- EP3438933B1 EP3438933B1 EP17774248.3A EP17774248A EP3438933B1 EP 3438933 B1 EP3438933 B1 EP 3438933B1 EP 17774248 A EP17774248 A EP 17774248A EP 3438933 B1 EP3438933 B1 EP 3438933B1
- Authority
- EP
- European Patent Office
- Prior art keywords
- light spot
- light
- moving
- moving amount
- spot
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 claims description 16
- 238000006243 chemical reaction Methods 0.000 claims description 10
- 238000003384 imaging method Methods 0.000 claims description 8
- 238000005259 measurement Methods 0.000 description 34
- 238000000034 method Methods 0.000 description 28
- 238000012545 processing Methods 0.000 description 16
- 238000010586 diagram Methods 0.000 description 10
- 239000013598 vector Substances 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000004297 night vision Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 239000000470 constituent Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000004438 eyesight Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/62—Extraction of image or video features relating to a temporal dimension, e.g. time-based feature extraction; Pattern tracking
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60Q—ARRANGEMENT OF SIGNALLING OR LIGHTING DEVICES, THE MOUNTING OR SUPPORTING THEREOF OR CIRCUITS THEREFOR, FOR VEHICLES IN GENERAL
- B60Q1/00—Arrangement of optical signalling or lighting devices, the mounting or supporting thereof or circuits therefor
- B60Q1/02—Arrangement of optical signalling or lighting devices, the mounting or supporting thereof or circuits therefor the devices being primarily intended to illuminate the way ahead or to illuminate other areas of way or environments
- B60Q1/04—Arrangement of optical signalling or lighting devices, the mounting or supporting thereof or circuits therefor the devices being primarily intended to illuminate the way ahead or to illuminate other areas of way or environments the devices being headlights
- B60Q1/14—Arrangement of optical signalling or lighting devices, the mounting or supporting thereof or circuits therefor the devices being primarily intended to illuminate the way ahead or to illuminate other areas of way or environments the devices being headlights having dimming means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
Description
- The present invention relates to an environment recognition device, for example, an environment recognition device for light distribution control.
- In order to secure (support) safety on traffic such as at night, the importance of light distribution control that assists a driver in ensuring a visual field by appropriately controlling a headlight depending on the environment in front of a host vehicle has been attracting attention. In the light distribution control, the headlight is set to a high beam when there is no preceding vehicle and oncoming vehicle in front of the vehicle, and the headlight is set to a low beam when there is a preceding vehicle or an oncoming vehicle. In order to implement such light distribution control, a technique has been developed in that light from a vehicle (other vehicle) is detected using an in-vehicle camera mounted at the front of the vehicle and light of the host vehicle is switched to a low beam on the basis of the detected signal.
- Incidentally, various lights (also referred to as light spot or point source of light) such as a street light, a traffic light, and a light reflector exist at night as well as a headlight of an oncoming vehicle and a tail light of a preceding vehicle. Accordingly, another technique has been also developed in that the light from other than the vehicle and the headlight/tail light of the vehicle are discriminated to avoid, due to erroneous detection of the light other than the vehicle, erroneously switching the headlight of the host vehicle to the low beam when there is no vehicle. However, especially with respect to the light in a distant place, there has been a problem that it is generally difficult to distinguish the light of the vehicle from the light other than the vehicle from an image of one frame.
- In response to such a problem,
PTL 1 discloses a conventional technique in which an object candidate is extracted from an image obtained by an in-vehicle imaging device using a pattern shape stored in advance, a changing amount prediction of the extracted object candidate is calculated on the basis of a method of predictively calculating a changing amount of an object differently set for each region among a plurality of regions into which the obtained image is divided, detected host vehicle behavior information, and a position of the object candidate, the object is tracked on the basis of a predicted position of the object, and it is determined whether the object (light spot) is a stationary object on a road on the basis of the result of tracking the object. - More specifically, the in-vehicle environment recognition device and the in-vehicle environment recognition system disclosed in
PTL 1 calculate a predicted position of the object candidate (candidate for point source of light) using the host vehicle behavior and the like, and tracks the candidate for the point source of light that exists at a location similar to the predicted position. In this technique, when tracking is established at a location similar to a result of the predictive calculation of the changing amount in each frame in the case of assuming a height on a road surface, it is assumed that the candidate for the point source of light is likely to be a stationary object on the road, thereby determining as a non-moving object. -
EP 2 575 078 A2
US 2010/289632 A1 discloses a system to display graphical images upon a windscreen of a vehicle including night vision including a transparent wind-screen head up display, a night vision system, and an enhanced Vision system system manager monitoring data from the night Vision system, analyzing the monitored data identifying critical information, and determining display requirements based upon the critical information.
PTL 1:JP 5320331 B2 - However, there are various heights of lights such as a street light, a traffic light, and a reflection of a sign, which are erroneously detected as vehicle lights. Accordingly, it is difficult to assume the height of such various lights on the road surface and to discriminate between a non-moving object (stationary light spot) and a moving object (moving light spot).
- The present invention has been conceived in view of the above-described problem, and an object of the present invention is to provide an environment recognition device capable of discriminating between a stationary light spot and a moving light spot even in a case where a height of a light spot concerned is unknown.
- In order to solve the problems described above, an environment recognition device according to the present invention includes a light-spot detection unit that is configured to detect a light spot from each image photographed at a plurality of points in time using an imaging device mounted on a host vehicle, and a moving-light-spot determination unit that is configured to determine, on the basis of variation in positions based on the host vehicle moving amount of light spots determined as identical among the light spots detected from the images photographed at the plurality of points in time, whether the light spot is a light spot moving relative to the ground. A reference generation unit is provided that is configured to generate reference for calculating the variation in positions of light spots using a position of the light spot detected from the image previously photographed by the imaging device. The reference generation unit is configured to calculate a moving amount of the light spot in a predetermined period using the position of the light spot detected from the image previously photographed by the imaging device, and is configured to generate the reference on the basis of the moving amount of the light spot in the predetermined period and a moving amount of the host vehicle in a period same as the predetermined period. Wherein the reference generation unit is configured to calculate the moving amount of the light spot by setting a height of the light spot to a predetermined value and deriving three-dimensional world coordinates from screen coordinates of the light spot on the image using perspective conversion. The reference generation unit is configured to generate the reference on the basis of a correlation between the moving amount of the light spot in the predetermined period and the moving amount of the host vehicle in the period same as the predetermined period. The correlation is represented by a correlation line determined from the moving amount of the light spot in the predetermined period and the moving amount of the host vehicle in the period same as the predetermined period. The correlation with the host vehicle moving amount is maintained without setting the true value to the height at the time of converting from the screen coordinates into the world coordinates.
- According to the present invention, it is determined whether the light spot is a light spot moving relative to the ground on the basis of the variation in the positions of the light spots detected at the plurality of points in time, whereby a stationary light spot and a moving light spot can be efficiently discriminated even in a case where a height of a light spot concerned is unknown.
- Problems, configurations, and effects other than those described above will be clarified in the following descriptions of the embodiments.
-
- [
FIG. 1] FIG. 1 is a block diagram illustrating an internal configuration of an environment recognition device according to a first embodiment of the present invention. - [
FIG. 2] FIG. 2 is a block diagram illustrating an internal configuration of a reference generation unit illustrated inFIG. 1 . - [
FIG. 3] FIG. 3 is a graph illustrating a relationship between a screen coordinate system of an image photographed by a camera and a world coordinate system obtained by converting the screen coordinate system into three dimensions. - [
FIG. 4] FIG. 4 is a graph illustrating processing of a light spot moving amount calculation unit illustrated inFIG. 2 , and exemplifies a case where a feature point is detected from an image. - [
FIG. 5] FIG. 5 is a diagram illustrating a moving amount of a light spot that can be obtained by the camera. - [
FIG. 6] FIG. 6 is a graph illustrating an exemplary stationary light spot tracked on a screen. - [
FIG. 7] FIG. 7 is a table illustrating an example of screen coordinates of a stationary light spot, X, Y, and Z coordinates as a result of converting the screen coordinates into world coordinates, and a moving amount on the world coordinates. - [
FIG. 8] FIG. 8 is a graph illustrating an example in which a relationship between a host vehicle moving amount and a light spot moving amount is plotted with a horizontal axis representing the host vehicle moving amount and a vertical axis representing the light spot moving amount. - [
FIG. 9] FIG. 9 is a graph illustrating an example in which two types of light spots are tracked and a relationship between a moving amount of the light spot on world coordinates and the host vehicle moving amount is plotted. - [
FIG. 10] FIG. 10 is a table illustrating an example of stored data stored in a light spot reference storage unit illustrated inFIG. 2 . - [
FIG. 11] FIG. 11 is a flowchart illustrating a process flow of a moving-light-spot determination unit illustrated inFIG. 1 . - [
FIG. 12] FIG. 12 is a table illustrating a state of stored data stored in the light spot reference storage unit at a point in time (time point t) at which the process of the moving-light-spot determination unit illustrated inFIG. 1 is completed. - [
FIG. 13] FIG. 13 is a diagram illustrating processing of a light distribution control unit illustrated inFIG. 1 , and illustrates an exemplary image photographed by the camera at the time point t. - [
FIG. 14] FIG. 14 is another diagram illustrating the processing of the light distribution control unit illustrated inFIG. 1 , and illustrates an exemplary image photographed by the camera at a time point t + 1. - [
FIG. 15] FIG. 15 is a table illustrating the processing of the light distribution control unit illustrated inFIG. 1 , and illustrates a state of stored data stored in the light spot reference storage unit at the point in time (time point t + 1) at which the process of the moving-light-spot determination unit is completed. - [
FIG. 16] FIG. 16 is a block diagram illustrating an internal configuration of an environment recognition device according to a second embodiment of the present invention. - [
FIG. 17] FIG. 17 is a block diagram illustrating an internal configuration of a distance measurement reliability determination unit illustrated inFIG. 16 . - [
FIG. 18] FIG. 18 is a table illustrating an example of stored data stored in a light spot position information storage unit illustrated inFIG. 17 . - [
FIG. 19] FIG. 19 is a graph illustrating a principle of a light spot height variation determination unit illustrated inFIG. 17 . - [
FIG. 20] FIG. 20 is a table illustrating an exemplary result of determining reliability from magnitude of variation in the light spot height variation determination unit illustrated inFIG. 17 . - Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.
- In the following descriptions, a case where, in order to control (light distribution control and headlight control) a light distribution state of a vehicle in which an environment recognition device according to the present invention is mounted, an object (light spot and point source of light) in front of the host vehicle is detected by a camera as an imaging device disposed to face the front to image the front of the vehicle will be described. Meanwhile, it should be noted that the present invention can also be applied to a case where an object behind or at the side of the host vehicle is detected.
-
FIG. 1 is a block diagram illustrating an internal configuration of an environment recognition device according to a first embodiment of the present invention. - As illustrated in the drawing, an
environment recognition device 1 basically includes animage acquisition unit 11, a light-spot detection unit 12, a host vehicle movingstate acquisition unit 13, areference generation unit 14, a moving-light-spot determination unit 15, and a lightdistribution control unit 16. - The
image acquisition unit 11 periodically (in a time series manner) obtains, from a camera (monocular camera) C1 installed at the front of a vehicle, for example, an image in front of the vehicle (host vehicle). - The light-
spot detection unit 12 detects, from the image obtained by theimage acquisition unit 11, information associated with a light spot included in the image. The light-spot detection unit 12 detects the light spot from the obtained image by, for example, extracting a high luminance region by performing binarization with an appropriate threshold value. Further, since theenvironment recognition device 1 detects both a preceding vehicle (tail light) and an oncoming vehicle (headlight), the light spot is detected by extracting, in addition to the high luminance region, a red region while color image information is used. - With respect to the extracted region, the light-
spot detection unit 12 obtains information associated with coordinates of a circumscribed rectangle of the region, an area of a high luminance pixel in the extracted region, luminance, and the like. With respect to the luminance, a representative value such as an average value, a maximum value, and a minimum value in the entire region is included. - The host vehicle moving
state acquisition unit 13 obtains information associated with a running state, which is obtained from various sensors for measuring a running state of the host vehicle, at every unit time (periodically). Any information may be used as the information to be obtained as long as it is a parameter capable of calculating a distance (moving amount) through which the host vehicle has moved, examples of which include a speed obtained from a vehicle speed sensor, a steering angle of a steering wheel obtained from a steering angle sensor, and a yaw rate obtained from a yaw rate sensor. - The
reference generation unit 14 generates reference for determining whether the detected light spot is moving relative to the ground (or whether the detected light spot is stationary relative to the ground) using the information associated with the light spot detected by the light-spot detection unit 12 at the plurality of points in time (that is, from each image photographed at the plurality of points in time) and the information associated with the running state of the host vehicle obtained by the host vehicle movingstate acquisition unit 13. - In the present embodiment, a moving amount on world coordinates is estimated from a movement of the light spot on screen coordinates of the image, and an average of relative values between the moving amount of the light spot (light spot moving amount) and the moving amount of the host vehicle (host vehicle moving amount) is generated as reference (to be described in detail later).
- The moving-light-
spot determination unit 15 determines whether each light spot in the image detected by the light-spot detection unit 12 is a moving light spot (light spot moving relative to the ground) or a stationary light spot (light spot stationary relative to the ground) on the basis of the reference generated by thereference generation unit 14. - The light
distribution control unit 16 determines light distribution control for determining whether to make a headlight of the host vehicle a high beam or a low beam on the basis of the determination result (whether the detected light spot is a moving light spot or a stationary light spot) of the moving-light-spot determination unit 15. IN the light distribution control here, the headlight of the host vehicle is set to a high beam normally, and when a tail light of a preceding vehicle, a headlight of an oncoming vehicle, or the like is detected, the headlight of the host vehicle is switched to a low beam so as not to bother a driver of the other vehicle. Therefore, the lightdistribution control unit 16 sets the light distribution state to a low beam when there is at least one light spot determined as a moving light spot by the moving-light-spot determination unit 15. On the other hand, when there is no light spot determined as a moving light spot by the moving-light-spot determination unit 15 (in other words, only when all light spots are determined as stationary light spots by the moving-light-spot determination unit 15), switching from the low beam to the high beam is permitted. - Next, the
reference generation unit 14, the moving-light-spot determination unit 15, and the lightdistribution control unit 16, which are characteristic parts of theenvironment recognition device 1 according to the present invention, will be described in detail with reference toFIGS. 2 to 14 . - As illustrated in
FIG. 2 , thereference generation unit 14 mainly includes a lightspot tracking unit 21, a light spot movingamount calculation unit 22, a host vehicle movingamount calculation unit 23, and a light spotreference storage unit 24. - The light
spot tracking unit 21 determines an identical light spot by matching with respect to each of a plurality of light spots detected by the light-spot detection unit 12 (in an image photographed at) a certain time point t and (in an image photographed at) the next time point t + 1. Various methods of the matching may be used as long as a moving locus of the light spot can be obtained. For example, when an image acquisition cycle is short, a method of matching the light spot at the closest position on the image may be used. Further, there is another method of matching in which a moving position is predicted on the basis of a moving direction and a moving distance on the image obtained from past tracking processing and the light spot closest to the predicted position is matched. Furthermore, as disclosed inPTL 1, there is still another method of matching in which the moving amount of the next time point on the world coordinates is determined on the basis of a moving state of a host vehicle, a predicted position is determined by converting the moving amount into a position on the image using perspective conversion, and the light spot closest to the predicted position is matched. - The light spot moving
amount calculation unit 22 estimates the moving amount of the light spot on the world coordinates on the basis of the moving locus of the light spot (light spot determined as identical) on the image obtained by a tracking result of the lightspot tracking unit 21. - A method of estimating the light spot moving amount by the light spot moving
amount calculation unit 22 will be described with reference toFIG. 3 . A screen coordinate system of an image photographed by the in-vehicle camera C1 is denoted by 31 inFIG. 3 , which is represented by two-dimensional coordinates of a horizontal axis u denoted by 33 and a vertical axis v denoted by 34. A world coordinate system obtained by converting the screen coordinatesystem 31 into three dimensions is denoted by 32 inFIG. 3 , which is represented by a horizontal axis X denoted by 36, a vertical axis Y denoted by 37, and a depth axis Z denoted by 38. -
- The focal length and the vanishing point coordinates are values uniquely determined as internal parameters of the camera. The moving distance on the assumption of a height can be estimated by converting the screen coordinates (system) for each time point to the world coordinates (system) using Formula (1).
- In Formula (1), Y1 representing the height of the point on the world coordinates cannot be determined from the position of the light spot on the image detected by the light-
spot detection unit 12. This is assumed to be an appropriate predetermined value, such as 1, to perform calculation. The reason for performing calculation by assuming Y1 will be described in detail later. - The light spot moving
amount calculation unit 22 converts the coordinates of the light spot tracked by the lightspot tracking unit 21 on the image into the world coordinates and calculates the distance on the world coordinates, thereby calculating the moving amount between the tracked light spots. -
FIG. 4 is a graph illustrating processing of the light spot movingamount calculation unit 22. On a screen coordinate system denoted by 41, light spots on the image are detected by the lightspot tracking unit 21 in an order of p401, p402, p403, p404, and p405. Although these positions are respectively detected in images at different time points, they are illustrated on the same image for convenience of explanation. The positions on the screen coordinate system are converted into positions p406, p407, p408, p409, and p410 on the world coordinate system denoted by 42 using Formula (1). Once each position on the world coordinates is determined, the moving amount can be determined from the coordinates before and after the light spot. - The host vehicle moving
amount calculation unit 23 calculates, from the information associated with the running state of the host vehicle obtained by the host vehicle movingstate acquisition unit 3, the moving amount of the host vehicle in the period same as the period in which the moving amount is calculated by the light spot movingamount calculation unit 22. - The method of calculating the moving amount of the host vehicle depends on the content of information that can be obtained by the host vehicle moving
state acquisition unit 13. In this case, a vehicle speed ν (m/sec) and a yaw rate ρ (deg/sec) for each cycle Δt (sec) can be obtained. -
- ΔX: movement amount in X direction
- ΔY: movement amount in Z direction
(Mathematical Formula 2) - The light spot moving
amount calculation unit 22 and the host vehicle movingamount calculation unit 23 store the calculated moving amounts (light spot moving amount and host vehicle moving amount) together with a time index in the light spot movingamount calculation unit 22 and the host vehicle movingamount calculation unit 23 so that the moving amounts can be referred to using time. - The light spot
reference storage unit 24 stores the reference generated on the basis of the light spot moving amount calculated by the light spot movingamount calculation unit 22 and the host vehicle moving amount calculated by the host vehicle movingamount calculation unit 23. - Here, a principle of the reference stored in the light spot
reference storage unit 24 and a method of generating the reference will be described. -
FIG. 5 is a diagram illustrating the moving amount of the light spot that can be obtained by the camera C1, which illustrates the image obtained by the in-vehicle camera C1 mounted on a running vehicle. In the drawing, light spots of a tail light of a precedingvehicle 59 are denoted by 51 and 52, a light spot of a street light on the roadside is denoted by 53, and a light spot of a light reflector on the roadside is denoted by 54. It is assumed that the image is photographed at the time point t, and movement destinations (directions and sizes) of the light spots 51, 52, 53, and 54 on the image photographed at the time point t + 1 in the next cycle are denoted bymovement vectors - Assuming that the moving amount of the host vehicle between the time point t and the time point t + 1 is Lown and the moving amount of the preceding
vehicle 59 is L59, a moving amount L55 of thevector 55 in the world coordinates conversion, a moving amount L56 of the vector 56 in the world coordinates conversion, a moving amount L57 of thevector 57 in the world coordinates conversion, and a moving amount L58 of thevector 58 in the world coordinates conversion are each expressed by the following Formula (3). - That is, the moving amount of the
light spot 53 of the street light or thelight spot 54 of the light reflector, which is a stationary object, coincides with the moving amount Lown of the host vehicle. Meanwhile, the moving amount of the light spots 51 and 52 of the light of the preceding vehicle, which are moving objects, is an amount obtained by adding or subtracting the moving amount of the object (moving object) to or from the moving amount Lown of the host vehicle. - Accordingly, when the moving amount of the detected light spot on the world coordinates can be actually measured, it can be discriminated whether the light spot is moving or stationary. Here, when Formula (1) to be used for the calculation in the light spot moving
amount calculation unit 22 is referred to, the height Y1 is required to convert the screen coordinates (u1, v1) into the world coordinates (X1, Y1, Z1). However, since information associated with the height of the light spot cannot be obtained from the screen coordinates of one shot, an absolute value of the moving amount cannot be calculated. - Meanwhile, when Formula (1) is referred to, the height Y1 of the converted object is a value effective for a multiplication in the formula (upper formula of Formula (1)) for calculating the depth Z1 on the world coordinates. Therefore, even when an accurate height cannot be obtained, the ratio between the moving amount calculated by assuming the height to an appropriate value and the moving amount calculated with the accurate height is constant.
-
FIG. 6 illustrates an exemplary stationary light spot tracked on a screen (image). For convenience of explanation, in this case, the positions of the light spots in a plurality of frames are denoted by reference signs p601 to p607 in the same image.FIG. 7 illustrates an example of screen coordinates of the stationary light spots p601 to p607, X, Y, and Z coordinates as a result of converting the screen coordinates into the world coordinates using Formula (1), and the moving amount on the world coordinates. - The reference signs of the light spots p601 to p607 illustrated in
FIG. 6 correspond to acolumn 701 inFIG. 7 . The Z and X coordinates on the world coordinates are calculated on the basis of the coordinates on the screen ofcolumns FIG. 7 ,columns columns columns column 706, thecolumn 709, and thecolumn 712 are calculated from the Z coordinate and the X coordinate of the light spot of a continuous time series. - It is assumed that, on the basis of the moving amount calculated in
FIG. 7 , the height Y = 1.0 is a true value, and the calculated moving amount coincides with the actual moving amount of the light spot in the case of Y = 1.0. As a result of this,FIG. 8 is a graph in which a relationship between the host vehicle moving amount and the light spot moving amount is plotted with a horizontal axis representing the host vehicle moving amount and a vertical axis representing the light spot moving amount. - A correlation equation for the plot of the light spot moving amount and the host vehicle moving amount when the height Y of the light spot is 1.0 is illustrated in L802. As it coincides, the inclination (correlation coefficient) of the correlation equation (correlation line) is 1. With respect to a correlation equation L801 in which the height Y is set to 2.0 and a correlation equation L803 in which the height Y is set to 0.5 as well, a correlation coefficient is around 1, which shows correlation. Accordingly, it is considered that the correlation with the host vehicle moving amount is maintained without setting the true value to the height at the time of converting from the screen coordinates into the world coordinates.
- From the above, by observing the variation of the ratio of the light spot moving amount based on the host vehicle moving amount (or, to the contrary, ratio of the host vehicle moving amount based on the light spot moving amount), it can be determined as a stationary object (stationary light spot) (in other words, stationary object in which the moving amount thereof coincides with that of the host vehicle (moving amount in the opposite direction)) stationary relative to the ground when the variation is small, and as a moving object (moving light spot) (moving object in which the moving amount thereof is different from that of the host vehicle) moving relative to the ground when the variation is large.
-
FIG. 9 illustrates an example in which two types of light spots are tracked and a relationship between the light spot moving amount on the world coordinates and the host vehicle moving amount is plotted. In the drawing, distribution of the moving amount of a light spot A relative to the host vehicle moving amount is denoted by 901, and distribution of the moving amount of a light spot B relative to the host vehicle moving amount is denoted by 904. A correlation equation relative to thedistribution 901 is denoted by L902, and magnitude of variation relative to the correlation equation L902 is denoted by D903. A correlation equation relative to thedistribution 904 is denoted by L905, and magnitude of variation relative to the correlation equation L905 is denoted by D906. - Here, the light spot having a small variation relative to the correlation equation like the light spot A may be defined as a stationary light spot, and the light spot having a large variation relative to the correlation equation like the light spot B may be defined as a moving light spot.
- The reference stored in the light spot
reference storage unit 24 illustrated inFIG. 2 is reference for describing the correlation between the host vehicle moving amount and the light spot moving amount. Therefore, the light spotreference storage unit 24 stores, as reference for each time point, a ratio obtained by dividing the light spot moving amount calculated by the light spot movingamount calculation unit 22 for each time point by the host vehicle moving amount in the same period calculated by the host vehicle movingamount calculation unit 23, as expressed by the following Formula (4). -
FIG. 10 illustrates an example of stored data stored in the light spotreference storage unit 24. Acolumn 1001 indicates a time point at which the reference is calculated. Acolumn 1002 indicates an ID for uniquely specifying the light spot for calculating the reference. Acolumn 1003 indicates the number of histories, which represents the number of tracking the light spot by the lightspot tracking unit 21. Acolumn 1004 indicates the reference, which is calculated from the moving amount of the light spot obtained at a current time point and the host vehicle moving amount during the same period using Formula (4) mentioned above. - Next, the moving-light-
spot determination unit 15 illustrated inFIG. 1 determines, as described above, whether each light spot in the image detected by the light-spot detection unit 12 is a moving light spot or a stationary light spot on the basis of the reference generated by thereference generation unit 14. - For example, a case where reference (light spot moving amount/host vehicle moving amount) Bn in a period (cycle) of a time N + 1 from the time point t to the time point t + N with respect to a certain light spot in the image is stored in the light spot
reference storage unit 24 of thereference generation unit 14 will be described as an example. - First, the moving-light-
spot determination unit 15 calculates an average value m in period of the time N + 1 (hereinafter may be referred to as reference average value) with respect to the reference Bn (Bn: reference for time point n) of the light spot determined as identical, as expressed by the following Formula (5). -
- As described with reference to
FIG. 9 , when the variance V of the reference is small, the ratio of the light spot moving amount relative to the host vehicle moving amount is constant as indicated by 901, which is on the correlation line L902. On the other hand, when the variance V of the reference is large, variation of the ratio of the light spot moving amount relative to the host vehicle moving amount increases, and variation relative to the correlation line L905 also increases, as indicated by the distribution D906. - In order to discriminate this difference, a threshold value is set in advance, and the light spot is determined as a stationary light spot when the variance V of the reference is less than the threshold value while the light spot is determined as a moving light spot when the variance V is equal to or more than the threshold value.
-
FIG. 11 illustrates a process flow performed by the moving-light-spot determination unit 15. The process flow performed by the moving-light-spot determination unit 15 will be described in a sequenced manner usingFIG. 10 . In this case, it is assumed that a current time is t and a threshold value for distinguishing the stationary light spot and the moving light spot is 0.5. - First, in S111, a light spot counter is initialized to 1.
- In S112, the reference average value m of a light spot i is calculated. Data associated with the light spot i is extracted from the light spot
reference storage unit 24 illustrated inFIG. 10 , and the average value is referred to. As an initial value of the light spot counter is 1, data with a light spot ID "1" in thecolumn 1002 is referred to. The data with the light spot ID "1" at the current time point t is in arow 1005, and the number of histories is "5". Therefore, five pieces of data with the light spot ID "1" including the above-mentioned data are sequentially referred to from the top of the data rows, thereby obtaining the reference. In the example illustrated inFIG. 10 , five pieces of data in therows - In S113, the reference variance V of a light spot i is calculated. In a similar manner to the average value calculation in S112, the data with the light spot ID "1" is obtained and the variance V = 3.87 is calculated using Formula (6) mentioned above as well as the average value m calculated in S112.
- In S114, the variance V of the light spot i calculated in S113 is compared with a predetermined threshold value. When the variance V is smaller than the threshold value, the light spot i is determined as a stationary light spot, and the process proceeds to S115. When the variance V is larger than the threshold value, the light spot i is determined as a moving light spot, and the process proceeds to S116. In the case of the light spot ID "1", the variance V = 3.87 is larger than the threshold value 0.5. Therefore, it is determined as a moving light spot and the process proceeds to S116.
- In S115, when it is determined as a stationary light spot, a moving light spot flag of the light spot i is set to 0.
- In S116, when it is determined as a moving light spot, the moving light spot flag of the light spot i is set to 1.
- In S117, the light spot counter i is advanced by 1. In this case, the light spot ID is set to "2".
- In step S118, it is determined whether all the light spots at the current time point t have been processed. When it is determined that all the light spots at the current time point t have been processed, all determination processing with respect to all the light spots existing at the time point is considered to have been completed, and the processing of the moving-light-
spot determination unit 15 is terminated. On the other hand, when it is determined that all the light spots at the current time point t have not been processed, the process returns to step S112, and the subsequent processing after S112 is also executed. In this case, as the light spot ID is "2", the process returns to S112, and data inrows reference storage unit 24 illustrated inFIG. 10 , thereby executing average value calculation processing and the like. Similarly, data inrows reference storage unit 24 illustrated inFIG. 10 , thereby executing the average value calculation processing and the like. - The processing result of the moving-light-
spot determination unit 15 is stored in the light spotreference storage unit 24 of thereference generation unit 14 together with the reference. -
FIG. 12 illustrates a state of stored data stored in the light spotreference storage unit 24 at the point in time when the process of the moving-light-spot determination unit 15 is completed. In addition to the table illustrated inFIG. 10 , the variance for each light spot and the moving light spot flag after the threshold value determination processing are added to acolumn 1205 and acolumn 1206, respectively. - By this processing, history data of the reference of the light spot existing at the current time point is referred to, and the variance thereof is calculated, whereby the moving-light-
spot determination unit 15 can determine whether each light spot in the image is a moving light spot or a stationary light spot. - Subsequently, the light
distribution control unit 16 illustrated inFIG. 1 determines, as described above, a light distribution state of the headlight of the host vehicle on the basis of the determination result (whether the detected light spot is a moving light spot or a stationary light spot) of the moving-light-spot determination unit 15. That is, the light distribution state is set to the low beam when it is determined as a moving light spot by the moving-light-spot determination unit 15 and there is at least one light spot having a moving light spot flag set to "1". On the other hand, when there is no light spot determined as a moving light spot by the moving-light-spot determination unit 15 (in other words, when all light spots are determined as stationary light spots by the moving-light-spot determination unit 15) and the moving light spot flags of all light spots are set to "0", switching from the low beam to the high beam is permitted. - This process of the light
distribution control unit 16 will be specifically described with reference toFIGS. 12 to 15 . -
FIG. 13 is animage 131 photographed by the in-vehicle camera C1 at the time point t, and a vehicle (oncoming vehicle) 132 and alight reflector 135 are reflected in theimage 131. In theimage 131, it is assumed that the detected light spots arelight spots vehicle 132, and alight spot 136 of thelight reflector 134. A result indicating how these light spots are determined by the moving-light-spot determination unit 15 is illustrated inFIG. 12 described above. Thelight spot 133 inFIG. 13 is the light spot with the light spot ID "1" inFIG. 12 , thelight spot 134 inFIG. 13 is the light spot with the light spot ID "2" inFIG. 12 , and thelight spot 136 inFIG. 13 is the light spot with the light spot ID "3" inFIG. 12 . When the light spot existing at the time point t is referred to inFIG. 12 , the moving light spot flag with the light spot ID "1" in arow 1207 and the moving light spot flag with the light spot ID "2" in arow 1208 are set to "1". Therefore, it is determined that the light of the vehicle exists, and in this case, the lightdistribution control unit 16 controls (switches) the light distribution state of the host vehicle to the low beam. - Besides,
FIG. 14 is animage 141 photographed by the in-vehicle camera C1 at the time point t + 1, and alight reflector 142 is reflected in theimage 141. In theimage 141, it is assumed that the detected light spot is alight spot 143 of thelight reflector 142.FIG. 15 illustrates a state of stored data stored in the light spotreference storage unit 24 reflecting the result determined by the moving-light-spot determination unit 15 with respect to this light spot. Thelight spot 143 inFIG. 14 is the light spot with the light spot ID "3" inFIG. 15 . When the light spot existing at the time point t + 1 is referred to inFIG. 15 , the moving light spot flag with the light spot ID "3" in arow 1501 is set to "0". Since this light spot is determined as a stationary light spot, in this case, the lightdistribution control unit 16 controls the light distribution state of the host vehicle to the high beam. - In this manner, as the moving-light-
spot determination unit 15 determines whether the light spot included in the image is a moving light spot or a stationary light spot, even in a case where a light spot is detected, when it is determined that the detected light spot is not a moving light spot (that is, not a light of a moving vehicle), the lightdistribution control unit 16 sets the light distribution state to the high beam, whereby erroneous control in which the light spot other than the vehicle, such as a street light and a light reflector, is erroneously detected and the light distribution state is set (switched) to the low beam can be prevented. - In the embodiment described above, the exemplary case where the existence/non-existence of the moving light spot is determined and the light distribution state is switched has been described. Here, when it is determined that the high beam/low beam of the host vehicle is switched according to the determination of the light
distribution control unit 16, control for delaying switching for several frames may be added. As a result, when an erroneous detection such as instantaneous presence of a moving object occurs, improper operation such as frequent switching of the light distribution state can be prevented. - As can be understood from the descriptions above, with the
environment recognition device 1 according to the present embodiment, it is determined whether the light spot is a light spot moving relative to the ground on the basis of the variation in the positions of the light spots detected at the plurality of points in time, whereby a stationary light spot and a moving light spot can be efficiently discriminated even in a case where a height of a light spot concerned is unknown (that is, even in a case where the actual accurate moving amount of the light spot cannot be measured). -
FIG. 16 is a block diagram illustrating an internal configuration of an environment recognition device according to a second embodiment of the present invention. - An
environment recognition device 2 according to the second embodiment illustrated inFIG. 16 is a device in which, mainly, adistance measurement unit 17 and a distance measurementreliability determination unit 18 are added to theenvironment recognition device 1 according to the first embodiment illustrated inFIG. 1 . Specifically, a means capable of measuring a distance to a light spot reflected on a camera image is added to the configuration according to the first embodiment, and when reliability of the measured distance is high, a light spot moving amount calculated by areference generation unit 14 is replaced with a light spot moving amount calculated on the basis of the distance measured by thedistance measurement unit 17, which is a configuration added to the configuration according to the first embodiment. Other configurations are substantially the same. Therefore, the configurations same as those in the first embodiment will be denoted by similar reference signs and detailed descriptions thereof will be omitted. Hereinafter, only the differences will be described in detail. - In the following descriptions, a case where the distance to the light spot is measured using a stereo camera including a plurality of cameras will be described. Meanwhile, it should be noted that any method may be used as long as the distance to the light spot can be measured.
- In the present embodiment, each of
image acquisition units distance measurement unit 17 obtains a parallax image from the image photographed by each camera C1 and C2 included in the stereo camera, thereby measuring the distance to the light spot reflected on the image. The method of calculating the parallax image is a conventionally known method, and the details thereof will be omitted. In the matching of right and left images (pattern matching), features on the image are used, whereby accuracy of the distance (measurement) is difficult to obtain at nighttime or the like during which the features on the image are difficult to obtain. Although the accuracy of the distance is high when an object is in the vicinity, it is difficult to determine how close the object should exist to obtain the high accuracy as it depends on conditions such as peripheral lightness. - In view of the above, the reliability of the measurement of the distance to the light spot is measured by the distance measurement
reliability determination unit 18, and only when the reliability is high, the moving amount (light spot moving amount) based on the distance measured by thedistance measurement unit 17 is calculated and the calculated light spot moving amount is replaced with the light spot moving amount calculated by thereference generation unit 14 to be used. - More specifically, as illustrated in
FIG. 17 , the distance measurementreliability determination unit 18 mainly includes a light spot positioninformation acquisition unit 171, a light spot positioninformation storage unit 172, a light spot heightvariation determination unit 173, and a light spot movingamount calculation unit 174. - The light spot position
information acquisition unit 171 obtains information associated with the distance (position) of the light spot measured by thedistance measurement unit 17. In this case, the distance on world coordinates (position coordinates) (X, Y, Z) measured by thedistance measurement unit 17 with respect to each light spot extracted by a light-spot detection unit 12 illustrated inFIG. 1 is obtained. The light spot positioninformation acquisition unit 171 refers to a light spotreference storage unit 24, which is a constituent element of thereference generation unit 14 illustrated inFIG. 2 , and stores the obtained distance information in the light spot positioninformation storage unit 172 in such a manner that the same light spot ID is set to the same light spot. - The light spot position
information storage unit 172 stores world coordinate position information of the light spot obtained by the light spot positioninformation acquisition unit 171 in association with the light spot ID tracked by thereference generation unit 14, and an example of the stored data is illustrated inFIG. 18 . The information stored in the light spotreference storage unit 24 illustrated inFIGS. 10 and12 can be collated with the same light spot on the basis of a time point in acolumn 1801 and a light spot ID in acolumn 1802 in data illustrated inFIG. 18 . - The light spot height
variation determination unit 173 determines stability of the distance information measured by thedistance measurement unit 17 from the variation of the position information of the light spot stored in the light spot positioninformation storage unit 172. - A principle of the light spot height
variation determination unit 173 will be described with reference toFIG. 19 . In the drawing, an image illustrating an exemplary movement of the position of the light spot is denoted by 191. Further, a position of the light spot at a time point t0 is denoted by p195, a position of the light spot at a time point tN is denoted by p196, and positions of the light spot therebetween are also illustrated in the drawing. - Although these are the same light spots as the light spots tracked by a light spot tracking unit 21 (see
FIG. 2 ), which are originally the light spots extracted on an image at different time points, they are illustrated on thesame image 191 for convenience of explanation. - The positions P195 and p196 of the light spot on the
image 191 and the position of the light spot on the world coordinates between the time points are obtained by thedistance measurement unit 17. At that time, a height Y of the light spot is also obtained. When the reliability of the distance measurement accuracy of the stereo camera is high, a light spot height is considered to be constant on the assumption that the road is flat and there is no slope, and even when there is a slight inclination, it is considered that the height measured by the stereo camera is not so different in a short period between the observation time points. The light spot heightvariation determination unit 173 of the distance measurementreliability determination unit 18 uses this and assumes that the distance measurement accuracy of thedistance measurement unit 17 is high when the variation of the height of the same light spot is small, and the distance measured by thedistance measurement unit 17 is used for the calculation of the light spot moving amount used in thereference generation unit 14. In agraph 192 inFIG. 19 , a horizontal axis represents time, and time-series variation of the height of the tracked light spot in theimage 191 is illustrated. Here, the variation of the height of the light spot is high in a period T193 illustrated in thegraph 192, and the variation of the height of the light spot is low in a period T194. Accordingly, it is assumed that the reliability of the distance measurement is low in the period T193, and the distance measured by thedistance measurement unit 17 during this period is not used. - Further, it is assumed that the reliability of the distance measurement is high in the period T194, and the distance measured by the
distance measurement unit 17 during this period is used. - Here, as a method of determining magnitude of the variation of the light spot height, variance of the heights of the past several frames including the current time point is calculated, and when the variance is smaller than a preset threshold value, it is assumed that the variance of the heights is stable and the distance information obtained in the frame is stable, and a reliability flag is set to "1".
FIG. 20 is an exemplary result of determining the reliability from the magnitude of the variation in the light spot heightvariation determination unit 173 with respect to the data illustrated inFIG. 18 . Y standard deviation, which is an index representing the variation of the height Y, is calculated and stored in acolumn 2001 inFIG. 20 on the basis of thecolumns 1801 to 1806 inFIG. 18 . In this case, the Y standard deviation in thecolumn 2001 is calculated from the height of the light spot in three past times as an example. Further, in the present embodiment, it is assumed that the reliability is high when the standard deviation is less than 10, and the reliability is low when the standard deviation is 10 or more, as a threshold value for determining the variation of the height Y. - The result of determining the reliability of the distance measurement of the light spot at each time point using the Y standard deviation calculated in the
column 2001 illustrated inFIG. 20 and the threshold value is illustrated in acolumn 2002. In the example illustrated inFIG. 20 , the light spot in which the Y standard deviation is less than 10 and the reliability of the distance measurement is determined as high is the light spot with the light spot ID "1" at the time point t in arow 2003, the light spots with the light spot ID "2" at the time point t inrows row 2006. - Although the number of data for calculating the standard deviation of the height Y is set to past three times and the threshold value of the Y standard deviation is set to 10 for convenience of explanation in the present embodiment, it is not necessarily limited to this.
- The light spot moving
amount calculation unit 174 illustrated inFIG. 17 calculates, when the reliability of the distance measurement is determined as high by the light spot heightvariation determination unit 173, the moving amount of the light spot (light spot having high distance measurement reliability) using the distance measured by thedistance measurement unit 17. - Here, in order to calculate the moving amount of the light spot between the time points, the light spot moving
amount calculation unit 174 uses the position of the light spot at the current time and the time of one time point period before the current time. Therefore, when the reliability flags of the light spot positions both at the current time and the time of one time period before the current time are set to "1", the moving amount of the light spot can be calculated using the distance measured by thedistance measurement unit 17. In the example illustrated inFIG. 20 , the moving amount is calculated in the cases with the light spot ID "2" in therows -
- ΔXt: movement amount in X direction at time point t
- Δ2t: movement amount in Z direction at time point t
(Mathematical Formula 7) - The light spot moving
amount calculation unit 174 replaces the calculated light spot moving amount with the light spot moving amount calculated by the reference generation unit 14 (in particular, light spot movingamount calculation unit 22 inFIG. 2 ), and thereference generation unit 14 calculates the subsequent reference using the light spot moving amount calculated by the light spot movingamount calculation unit 174. - Then, as described above, a moving-light-
spot determination unit 15 determines whether each light spot in the image detected by the light-spot detection unit 12 is a moving light spot or a stationary light spot on the basis of the reference generated by thereference generation unit 14, and a lightdistribution control unit 16 determines a light distribution state of a headlight of the host vehicle on the basis of a determination result (whether the detected light spot is a moving light spot or a stationary light spot) of the moving-light-spot determination unit 15. - In this manner, with the use of the
environment recognition device 2 according to the present embodiment, discrimination accuracy between the stationary object (stationary light spot) and the moving object (moving light spot) can be further improved by using the distance measurement result with high reliability. - As described above, it becomes possible to efficiently discriminate whether the detected light spot is a moving light spot or a stationary light spot by applying the
environment recognition devices - In the first and second embodiments, the exemplary case where the determination result of the moving object is used for the light distribution control, which is switching between the high beam and the low beam, has been described. Meanwhile, the determination result of the moving object can be also similarly used for shading control that controls a region irradiated with headlight, for example. In the shading control, a region in which light of a vehicle (other vehicle) exists is detected, and the region is transmitted to a control device, thereby controlling the irradiation of the headlight corresponding to the region. In this case, the irradiation of the headlight to the region can be limited by transmitting coordinates of the light spot with a moving light spot flag set to "1" to the control device.
- Note that the present invention is not limited to the first and second embodiments described above, and includes various modifications. For example, the above-described embodiments have been described in detail for convenience of explaining the present invention in a manner easy to understand, and are not necessarily limited to those having all the described configurations. A configuration of one embodiment may be partially replaced with a configuration of another embodiment, and the configuration of another embodiment may be added to the configuration of one embodiment. It is also possible to add, delete, and replace other configurations with respect to a part of the configuration of each embodiment.
- Moreover, each of the above-described configurations, functions, processing units, processing means, and the like may be partially or entirely implemented by hardware, for example, by designing them with an integrated circuit and the like. Each of the above-described configurations, functions, and the like may be implemented by software by interpreting and executing a program for implementing each function using a processor. Information such as a program, a table, and a file for implementing each function may be stored in a storage device such as a memory, a hard disk, and a solid state drive (SSD) or a recording medium such as an IC card, an SD card, and a DVD.
- Besides, control lines and information lines are indicated using what is considered to be necessary for explanation, and all the control lines and the information lines are not necessarily illustrated for products. In practice, it may be considered that almost all the configurations are mutually connected.
-
- 1
- environment recognition device (first embodiment)
- 2
- environment recognition device (second embodiment)
- 11
- image acquisition unit
- 12
- light-spot detection unit
- 13
- host vehicle moving state acquisition unit
- 14
- reference generation unit
- 15
- moving-light-spot determination unit
- 16
- light distribution control unit
- 17
- distance measurement unit
- 18
- distance measurement reliability determination unit
- 19
- image acquisition unit
- 21
- light spot tracking unit
- 22
- light spot moving amount calculation unit
- 23
- host vehicle moving amount calculation unit
- 24
- light spot reference storage unit
- 171
- light spot position information acquisition unit
- 172
- light spot position information storage unit
- 173
- light spot height variation determination unit
- 174
- light spot moving amount calculation unit
- C1
- camera
- C2
- camera
Claims (3)
- An environment recognition device, comprising:an image acquisition unit (11) that is configured to obtain an image from a camera which generates the image by photographing;a light-spot detection unit (12) that is configured to detect a light spot from each image photographed at a plurality of points in time using an imaging device mounted on a host vehicle;a host vehicle moving state acquisition unit (13) that is configured to receive the moving amount of the vehicle from sensors for periodically measuring a running state of the host vehicle;a reference generation unit (14) that is configured to generate reference for calculating the variation in positions of light spots using a position of the light spot detected from the image previously photographed by the imaging device,whereinthe reference generation unit (14) is configured to calculate a moving amount of the light spot in a predetermined period using the position of the light spot detected from the image previously photographed by the imaging device, and is configured to generate the reference on the basis of the moving amount of the light spot in the predetermined period and a moving amount of the host vehicle in a period same as the predetermined period; anda moving-light-spot determination unit (15) that is configured to determine, on the basis of the reference generated by the reference generation unit (14) and variation in positions of light spots determined as identical among the light spots detected from the images photographed at the plurality of points in time, whether the light spot is a light spot moving relative to a ground,characterized in that the reference generation unit (14) is configured to calculate the moving amount of the light spot by setting a height of the light spot to a predetermined value and deriving three-dimensional world coordinates from screen coordinates of the light spot on the image using perspective conversion, whereinthe reference generation unit (14) is configured to generate the reference on the basis of a correlation between the moving amount of the light spot in the predetermined period and the moving amount of the host vehicle in the period same as the predetermined period, and whereinthe correlation is represented by a correlation line determined from the moving amount of the light spot in the predetermined period and the moving amount of the host vehicle in the period same as the predetermined period,and wherein the correlation with the host vehicle moving amount is maintained without setting the true value to the height at the time of converting from the screen coordinates into the world coordinates.
- The environment recognition device according to claim 1, whereinthe reference generation unit (14) is configured to generate the reference from a ratio between the moving amount of the light spot in the predetermined period and the moving amount of the host vehicle in the period same as the predetermined period, andthe moving-light-spot determination unit (15) is configured to determine whether the light spot is a light spot moving relative to the ground on the basis of variation in the ratio.
- The environment recognition device according to claim 1, further comprising:a host vehicle moving state acquisition unit (13) that is configured to obtain information associated with a moving state of the host vehicle;a reference generation unit (14) that is configured to track a light spot determined as identical from the light spot detected from the image photographed at the plurality of points in time, is configured to calculate a moving amount of the light spot in a predetermined period, calculates a moving amount of the host vehicle in a period same as the predetermined period on the basis of the information obtained by the host vehicle moving state acquisition unit (13), and is configured to generate reference for calculating variation in positions of the light spot from a correlation between the moving amount of the light spot in the predetermined period and the moving amount of the host vehicle in the period same as the predetermined period; whereinvariation in positions of the light spot is calculated on the basis of the reference.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2016074561A JP6756507B2 (en) | 2016-04-01 | 2016-04-01 | Environmental recognition device |
PCT/JP2017/010054 WO2017169704A1 (en) | 2016-04-01 | 2017-03-14 | Environment recognition device |
Publications (3)
Publication Number | Publication Date |
---|---|
EP3438933A1 EP3438933A1 (en) | 2019-02-06 |
EP3438933A4 EP3438933A4 (en) | 2020-04-01 |
EP3438933B1 true EP3438933B1 (en) | 2023-10-04 |
Family
ID=59964265
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP17774248.3A Active EP3438933B1 (en) | 2016-04-01 | 2017-03-14 | Environment recognition device |
Country Status (3)
Country | Link |
---|---|
EP (1) | EP3438933B1 (en) |
JP (1) | JP6756507B2 (en) |
WO (1) | WO2017169704A1 (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPWO2019156087A1 (en) * | 2018-02-07 | 2021-04-01 | 株式会社小糸製作所 | Image processing equipment and vehicle lighting equipment |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007076378A (en) * | 2005-09-09 | 2007-03-29 | Nissan Motor Co Ltd | Lighting apparatus and method for vehicle |
JP5310235B2 (en) * | 2009-04-27 | 2013-10-09 | 株式会社豊田中央研究所 | On-vehicle lighting control device and program |
US8164543B2 (en) * | 2009-05-18 | 2012-04-24 | GM Global Technology Operations LLC | Night vision on full windshield head-up display |
JP5809785B2 (en) * | 2010-07-30 | 2015-11-11 | 日立オートモティブシステムズ株式会社 | Vehicle external recognition device and light distribution control system using the same |
CN103029621B (en) * | 2011-09-30 | 2016-04-06 | 株式会社理光 | Detect the method and apparatus of front vehicles |
JP5643877B2 (en) * | 2013-05-27 | 2014-12-17 | 株式会社小糸製作所 | Vehicle headlamp device |
JP5820843B2 (en) * | 2013-05-29 | 2015-11-24 | 富士重工業株式会社 | Ambient environment judgment device |
CN104424648B (en) * | 2013-08-20 | 2018-07-24 | 株式会社理光 | Method for tracing object and equipment |
-
2016
- 2016-04-01 JP JP2016074561A patent/JP6756507B2/en active Active
-
2017
- 2017-03-14 WO PCT/JP2017/010054 patent/WO2017169704A1/en active Application Filing
- 2017-03-14 EP EP17774248.3A patent/EP3438933B1/en active Active
Also Published As
Publication number | Publication date |
---|---|
JP6756507B2 (en) | 2020-09-16 |
JP2017187858A (en) | 2017-10-12 |
WO2017169704A1 (en) | 2017-10-05 |
EP3438933A1 (en) | 2019-02-06 |
EP3438933A4 (en) | 2020-04-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5938569B2 (en) | Advanced driver support system considering azimuth information and operation method thereof | |
US10627228B2 (en) | Object detection device | |
JP4420011B2 (en) | Object detection device | |
US8890951B2 (en) | Clear path detection with patch smoothing approach | |
US9047518B2 (en) | Method for the detection and tracking of lane markings | |
JP2013225295A5 (en) | ||
EP2928178B1 (en) | On-board control device | |
EP3410416B1 (en) | Image processing device, imaging device, mobile entity apparatus control system, image processing method, and program | |
JP6450294B2 (en) | Object detection apparatus, object detection method, and program | |
JP6794243B2 (en) | Object detector | |
KR20130007243A (en) | Method and system for warning forward collision using camera | |
US8559727B1 (en) | Temporal coherence in clear path detection | |
KR102359083B1 (en) | Device for detecting moving object and method thereof | |
Petrovai et al. | A stereovision based approach for detecting and tracking lane and forward obstacles on mobile devices | |
KR20180063524A (en) | Method and Apparatus for Detecting Risk of Forward Vehicle Using Virtual Lane | |
JP6263453B2 (en) | Momentum estimation device and program | |
JP6115429B2 (en) | Own vehicle position recognition device | |
KR20180047149A (en) | Apparatus and method for risk alarming of collision | |
EP3438933B1 (en) | Environment recognition device | |
JP2009276906A (en) | Travelling information providing device | |
KR102100047B1 (en) | Method for position recognition of vehicle using lane-end-point detection algorithm and method for evaluating performance of the same | |
JP6431299B2 (en) | Vehicle periphery monitoring device | |
JP6232883B2 (en) | Own vehicle position recognition device | |
JP7454685B2 (en) | Detection of debris in vehicle travel paths | |
JP2018088234A (en) | Information processing device, imaging device, apparatus control system, movable body, information processing method, and program |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
17P | Request for examination filed |
Effective date: 20181102 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
AX | Request for extension of the european patent |
Extension state: BA ME |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
DAV | Request for validation of the european patent (deleted) | ||
DAX | Request for extension of the european patent (deleted) | ||
RIC1 | Information provided on ipc code assigned before grant |
Ipc: G06T 7/20 20170101AFI20191107BHEP Ipc: G06K 9/00 20060101ALI20191107BHEP Ipc: B60Q 1/14 20060101ALI20191107BHEP Ipc: G06K 9/46 20060101ALI20191107BHEP |
|
A4 | Supplementary search report drawn up and despatched |
Effective date: 20200304 |
|
RIC1 | Information provided on ipc code assigned before grant |
Ipc: G06K 9/00 20060101ALI20200227BHEP Ipc: G06T 7/20 20170101AFI20200227BHEP Ipc: B60Q 1/14 20060101ALI20200227BHEP Ipc: G06K 9/46 20060101ALI20200227BHEP |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
17Q | First examination report despatched |
Effective date: 20210602 |
|
RAP3 | Party data changed (applicant data changed or rights of an application transferred) |
Owner name: HITACHI ASTEMO, LTD. |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R079 Ref document number: 602017074923 Country of ref document: DE Free format text: PREVIOUS MAIN CLASS: G06T0007200000 Ipc: G06V0020580000 Ref country code: DE Ref legal event code: R079 Free format text: PREVIOUS MAIN CLASS: G06T0007200000 Ipc: G06V0020580000 |
|
GRAP | Despatch of communication of intention to grant a patent |
Free format text: ORIGINAL CODE: EPIDOSNIGR1 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: GRANT OF PATENT IS INTENDED |
|
RIC1 | Information provided on ipc code assigned before grant |
Ipc: G06T 7/20 20170101ALI20230320BHEP Ipc: G06V 10/62 20220101ALI20230320BHEP Ipc: B60Q 1/14 20060101ALI20230320BHEP Ipc: G06V 20/58 20220101AFI20230320BHEP |
|
INTG | Intention to grant announced |
Effective date: 20230419 |
|
GRAS | Grant fee paid |
Free format text: ORIGINAL CODE: EPIDOSNIGR3 |
|
GRAA | (expected) grant |
Free format text: ORIGINAL CODE: 0009210 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE PATENT HAS BEEN GRANTED |
|
AK | Designated contracting states |
Kind code of ref document: B1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
REG | Reference to a national code |
Ref country code: GB Ref legal event code: FG4D |
|
REG | Reference to a national code |
Ref country code: CH Ref legal event code: EP |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R096 Ref document number: 602017074923 Country of ref document: DE |
|
REG | Reference to a national code |
Ref country code: IE Ref legal event code: FG4D |
|
REG | Reference to a national code |
Ref country code: LT Ref legal event code: MG9D |
|
REG | Reference to a national code |
Ref country code: NL Ref legal event code: MP Effective date: 20231004 |
|
REG | Reference to a national code |
Ref country code: AT Ref legal event code: MK05 Ref document number: 1618555 Country of ref document: AT Kind code of ref document: T Effective date: 20231004 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: NL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20231004 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: GR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20240105 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: IS Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20240204 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: LT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20231004 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: AT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20231004 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: ES Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20231004 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: LT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20231004 Ref country code: IS Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20240204 Ref country code: GR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20240105 Ref country code: ES Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20231004 Ref country code: BG Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20240104 Ref country code: AT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20231004 Ref country code: PT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20240205 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: DE Payment date: 20240331 Year of fee payment: 8 |